Shouichiro Iio
Shouichiro Iio
Shouichiro Iio
3 (2011) 224228
Shouichiro Iio1, Yusuke Katayama2, Fuminori Uchiyama3, Eiichi Sato4 and Toshihiko Ikeda1
1. Faculity of Eng., Shinshu University, 4-17-1 Wakasato, Nagano, JAPAN, 380-8553
2. Student of Shinshu University, 4-17-1 Wakasato, Nagano, JAPAN, 380-8553
3. Flowserve Niigata Worthington Co., Ltd., 1-32 Shinbashi, Kashiwazaki, Niigata, JAPAN, 945-0056
4. Faculty of Eng., Niigata Institute of Technology,1719 Fujihashi, Kashiwazaki, Niigata, JAPAN, 945-1195
© Science Press and Institute of Engineering Thermophysics, CAS and Springer-Verlag Berlin Heidelberg 2010
The aim of this investigation was to improve power performance of Savonius hydraulic turbine utilizing small
stream for electric generation. An attempt was made to increase the power coefficient of runner by the use of flat
shield plate placed upstream of the runner. The difference of the power coefficient is discussed in relation to
clearance between the runner and the bottom wall and the rotation direction of the runner. The flow field around
the runner was also examined visually to clarify influences of setting conditions on the power performance. From
this study it was found that the power coefficient is achieved for 0.47 by only using a flat shield plate, the increase
is up to 80% over the runner without the plate. Moreover, it is the proper condition that clearance ratio is 0.73 in
this study.
Received: 2011
Shouichiro IIO: Associate Professor
www.springerlink.com
2 J. Therm. Sci., Vol.20, No.3, 2011
wind collector. that before the runner installation. But through this expe-
The performance improvement of the Savonius runner riment, the pump speed of the circuit tunnel was kept
is an important challenge also in this article for practical constant. The flow velocity distribution in the test section
using of the hydraulic turbines. Until now, Nakajima et al. is uniform except for near the side and bottom walls in
first attempted to improve the efficiency by applying a the tunnel. The normal water level in the test section is
phase difference to the blades of the Savonius rotor (24). set as H0=0.4 m. The distance between the runner and the
This runner had 90-degrees phase difference in the blades bottom wall of the tunnel is varied 0.033 to 0.154 m, and
between both sides of the partition plate. Additionally, is defined as HC; accordingly the clearance ratio HC/DR is
they revealed that the power performance changes de- changed from 0.23 to 1.08. A shield plate is using a flat
pend on the distance between the runner and the bottom plate because of study of Ogawa et al. that estimated the
wall of channel, and the rotation direction of the run- shape of the shield plate(12), and also from point of cost
ner(25). The maximum power coefficient was reached at and maintenance.
CPmax=0.26 under the best installation condition for small
rivers. This method is more conventional than installing
guide vanes or others around the runner, and is expected
to improve the performance without losing the advantage
of the Savonius runner, i.e. the easy maintenance, dura-
bility and low cost. But it is still necessary to greatly im-
prove the power coefficient from the practical stand point
of view. Therefore Iio et al. investigated the performance
improvement by installing a flat plate (shield plate) up-
stream the runner considering installation cost. As a re-
sult, the power coefficient was reached at CPmax=0.47(26).
It was because the shield plate blocked the stagnation
flow toward the convex side of returning blade, and ac-
celerated the flow velocity toward concave side of ad-
vancing blade.
The goal of this investigation is to achieve higher
runner performance. But it was still not clarified the op-
Fig. 1 Test section
timum installation condition attaching the shield plate
when practically setting the turbine in the rivers.
The purpose of this experimental study is to clarify the
effect of the water surface and bottom wall on the power
performance with shield plate.
It is obviously recognized that the CPmax is enhanced in 0.0 0.5 1.0 1.5 2.0 2.5
λ
HC/DR>0.73 for CW and in HC/DR<0.73 for CCW. (a) CW rotation
Similar pattern can also be seen in the case without the 0.5
shield plate. For HC/DR=0.73, the power coefficient has CCW
the maximum value of CPmax=0.47 for CW with the 0.4
shield plate, it is improved up to 80%. From the result of
performance for CW with the shield plate, the water sur- 0.3
CP
hanced because of blocking effect of stagnation flow ing to the water surface for CCW.
toward a returning blade and of occurring negative
pressure upstream the returning blade. For CCW rotation Conclusions
as shown in Fig. 5(c) and (d), it can be easily recognized
that the flow over the shield plate and through between The influence of the water surface and the bottom wall
the shield plate and the upper plate is apparently different. on the power performance is almost similar to that of
For HC/DR=0.73, the flow toward the returning blade is runner without the shield plate. In particular, the flow
not observed. As a result, it is clear that the size of wake near the water surface has little influence on the power
region and the block effect of stagnation flow toward the performance for CW rotation, on the other hand, influ-
returning blade are most important factors to improve the ence of the bottom wall is small for CCW rotation.
power coefficient of Savonius runner with shield plate. The power coefficient is reached at 0.47 by only using
And the water surface and the bottom wall affect on the a flat shield plate, the increase of approximately 80%
power performance. Fig. 5(a) shows vortices from top over the runner without the plate. The rotation speed of
and bottom tip of the shield plate, therefore pressure drop the runner is almost enhanced by the shield plate. The
is remarkable. But Fig. (b) shows vortex from only top best orientation conditions of the runner with the shield
tip of the shield plate, so negative pressure behind the plate are as follows; the clearance ratio defined the dis-
shield plate is lower than that of Fig. (a); accordingly it is tance between the runner and the bottom wall of channel
not easy to draw the returning blade for CW rotation. Fig. to the runner diameter is 0.73.
(d) shows the impinging flow on the returning blade ow-
Acknowledgement
References